Skip to main content
Internet Archive's 25th Anniversary Logo

Full text of "Gametophytes and Embryo of Torreya Taxifolia"

See other formats

Early Journal Content on JSTOR, Free to Anyone in the World 

This article is one of nearly 500,000 scholarly works digitized and made freely available to everyone in 
the world by JSTOR. 

Known as the Early Journal Content, this set of works include research articles, news, letters, and other 
writings published in more than 200 of the oldest leading academic journals. The works date from the 
mid-seventeenth to the early twentieth centuries. 

We encourage people to read and share the Early Journal Content openly and to tell others that this 
resource exists. People may post this content online or redistribute in any way for non-commercial 

Read more about Early Journal Content at 
journal-content . 

JSTOR is a digital library of academic journals, books, and primary source objects. JSTOR helps people 
discover, use, and build upon a wide range of content through a powerful research and teaching 
platform, and preserves this content for future generations. JSTOR is part of ITHAKA, a not-for-profit 
organization that also includes Ithaka S+R and Portico. For more information about JSTOR, please 


Botanical Gazette 

MARCH, 1905 




John M. Coulter and W. J. G. Land. 

Torreya taxijolia Arnott 1 occurs in a narrow belt along the eastern 
side of the Apalachicola River, extending from the southern boundary 
of Georgia for about thirty miles southward (1). In April 1904 this 
region was visited by H. C. Cowles of this laboratory, among whose 
notes the following are of interest in this connection : 

My visit was to the northernmost colony, west of Chattahoochee village, and 
close to the Georgia line. The distribution lines on Chapman's map (1) would 
lead one to suppose that the tree is xerophytic and frequents the steep and dry 
eastern bluffs. I was much surprised to find that it was confined (in the Chatta- 
hoochee station at least) to the extremely mesophytic slopes of ravines, growing 
exclusively in the shade of trees, and in places that are continually moist, prefer- 
ably on slopes facing north. The northern and southern known limits of the 
tree are only about thirty miles apart, and the east-west range is much less. 
Furthermore, on account of the great economic value of the wood, and the 
familiarity of the tree to all the inhabitants of the region, the likelihood of finding 
other areas is very slight. 

It is associated with a remarkable and somewhat extensive group of northern 
mesophytic plants, and the conclusion is irresistible that Torreya is a northern 
plant of the most pronounced mesophytic tendencies, and to be associated with 

1 Unfortunately, Torreya was used at least three times as a generic name, in as 

many families, before 1838, the date of the publication of Torreya Arnott. Hence 

Arnott' s genus has been replaced by Tumion Raf., and our species becomes Tumion 

taxijolia Greene. In the present paper, however, the more familiar name is used for 




such forms as the beech-maple-hemlock forms of our northern woods, our most 
mesophytic type of association. Probably it never becomes a large tree, although 
farmers always cut the trees as soon as they become at all usable. Rarely were 
any found over 3o cm in diameter or more than 9 to i2 m high. It has remark- 
able capacity for vegetative reproduction, almost equaling the redwood in this 
respect. Many suckers issue from cut stumps, and even from fallen trunks; 
even rotten stumps show vigorous suckers, and it seems to be as tenacious of life 
as the poplar. Staminate trees appear to be the more numerous; however, 
when in blossom they are far more conspicuous than are the pistillate trees, and 
the conclusion as to proportion may not hold. 

Cowles secured collections of staminate and ovulate material 
April 4 and 5, and arranged for subsequent collections to be sent to 
the laboratory. These collections were made at intervals of approxi- 
mately two weeks up to October 21. The strobilus-bearing twigs 
were packed in damp cotton in tightly closed tin buckets and reached 
the laboratory in good condition in about five days. The length of 
the intervals between collections and between collecting and killing 
has prevented so close a series as is desirable at certain stages, but 
the gymnosperm periods are so long that a fair outline of morpho- 
logical events was secured. 


Strobili. — The staminate strobili occur in the axils of leaves of 
young shoots. Their appearance in October, on a shoot of the 
same season, is shown in fig. 1. At this time they are small and 
ovoid, solitary in the axil of each leaf, and by a curving of the short 
peduncles are displayed along the under surface of the shoot. The 
strobilus consists of a series of closely overlapping sterile bracts, in 
four vertical rows, completely enveloping the tip of the axis bearing 
numerous stamens. It is thus distinctly sterile below and fertile 
above, the sporophylls continuing the spiral succession developed 
by the bracts (figs. 6 and 7). 

The young strobili were first observed in July, and at that time 
no primordia of sporophylls were evident; but in August these were 
beginning to appear. Fig. 7 is from a longitudinal section through 
a strobilus at this period (August 12), showing the overlapping sterile 
bracts and the beginning of stamineal primordia. A remarkable 
development of the pith region of the axis below the staminate por- 


tion of the strobilus, with its investing vascular cylinder, may be 
observed, apparently an important storage region for the strobilus. 

Stamens. — At the stage of the stamineal primordia observed 
August 12, there is no evident differentiation of tissues (fig. 8), but 
early in September the young sporangia become distinct, the sporo- 
genous tissue being represented in each case by a single primary 
sporogenous cell. At this stage the further development of the 
young sporangia differs. Seven sporangia are seen in the primary 
sporogenous cell stage, three adaxial and four abaxial (fig. 10), 
radially arranged about the central axis as are the five or more spor- 
angia in the allied Taxus. The four abaxial sporangia develop in 
the usual way, and become the pendent abaxial sporangia character- 
istic of Torreya. The three adaxial sporangia, however, do not 
develop further, the primary sporogenous cell not dividing, and its 
nucleus showing signs of disorganization (fig. 9). The disorganiza- 
tion continues until it involves all the cells separating the three adaxial 
sporangia, which are thus replaced by a single large flattened cavity, 
which becomes a resin cavity. Intermediate stages were found in 
which the three sporogenous regions were still distinct (fig. 10). In 
fig. 11 (October 21) is shown a median longitudinal section through 
a stamen, in which the large resin cavity is seen above and the nor- 
mally developing sporogenous tissue below. As a result, the mature 
stamen of Torreya is characterized by four abaxial sporangia and a 
large adaxial resin cavity. 

In one instance the two lateral sporangia were smaller than the 
middle ones (fig. 10), suggesting a tendency to still greater reduction 
in the number of sporangia. Accordingly Pinus Laricio was exam- 
ined, and two resin cavities were found related to the two sporangia 
exactly as the two lateral sporangia of Torreya are to the middle 
ones ; and in early stages the tissue on the sites of these two resin cavi- 
ties of Pinus resembles that of sporangia. It is interesting to note 
that in T. californica Miss Robertson (8) found occasional stamens 
bearing six or seven mature sporangia, indicating that in these cases 
some or all of the usually abortive sporangia reached maturity. 

There is evident a tendency to reduce the number of functioning 
sporangia by abortion, a reduction that has proceeded farther in 
Pinus than in Torreya; and in the latter farther than in Taxus. That 


the cavities formed by the breaking down of young sporogenous and 
adjacent tissue should become resin cavities in conifers is to be 

It was only in the abortive sporangia of Torreya that stages were 
found indicating the history of the sporangium up to the primary 
sporogenous cell stage. For example, from fig. g it seems evident 
that there was a single hypodermal archesporial cell, which at this 
stage had given rise to two layers of wall cells and a single primary 
sporogenous cell. In the mature sporangium there are three or four 
wall layers. The epidermis contains numerous stomata, and its 
cells become prominent and have thickened walls. The sporangia 
are bluntly four-angled from pressure, and the epidermal cells at 
the angles are much larger than elsewhere. 

Male gametophyte. — We secured no stages between the very 
young sporangium and the shedding of the pollen, which occurs late 
in March and early in April. Miss Robertson (8) has observed 
the mother- cell stage and the formation of tetrads in T. calijomica. 

In germination the microspore of T. taxijolia cuts off no prothallial 
cell, a feature common to all the Coniferales except Podocarpeae 
and Abieteae. After the first division the generative and tube cells 
are distinct and separated by a very delicate wall or membrane. 
The tube nucleus is sometimes spherical, but more often amoeboid 

(fig. 12). 

Early in April the binucleate pollen grains, rich in starch, were 
found resting on the nucellus (fig. ig) ; and at the end of June micro- 
pyles were found full of pollen grains, most of which had sent out no 
tubes, but were still full of starch, nucleate, and apparently alive. 

As fertilization occurs about the middle of August, active pollen 
tubes occupy the sterile cap of the nucellus about four months, and 
their behavior is exceedingly variable. They may advance towards 
the embryo sac rapidly, reaching it at a very early stage, as early as 
June 21, when the endosperm consists of only sixteen to sixty-four 
parietally placed free nuclei (figs. 14 and 15); or they may advance 
slowly, being found at all stages of progress at the same date. They 
may advance so directly that the tube resembles a straight cleft 
through the nucellar tissue to the sac, or they may pursue a remark- 
ably devious course. In one case a tube passed directly half-way 


down the nucellar cap, then advanced spirally downward and out- 
ward until it reached the peripheral cells of the nucellus, and after 
destroying several of these it turned abruptly inward, penetrated the 
nucellus at the level of the archegonium, crossed the top of the endo- 
sperm, and discharged its contents into the archegonium on the farther 
side. In another instance, such a pollen tube had entered the inner 
integument and destroyed some of its cells before turning back into 
the nucellus. It is very common for the pollen tube to push into the 
sac at its free nuclear stage, making a deep invagination, often to the 
middle; from this pocket the tube turns back again into the nucellus. 
It is of interest to note in this connection that the usually solitary 
archegonium is never centrally placed in the endosperm, and the 
pollen tube enters the sac to one side of it (fig. 16). 

In all the wanderings of the pollen tube, the body cell, stalk 
nucleus, and tube nucleus are conspicuous. The division of the 
generative cell was not observed, but that it occurs very early in 
the development of the tube is evident. The body cell is relatively 
very large, with a conspicuous nucleus and investing cytoplasm, and 
was always found consorting closely with the stalk and tube nuclei 
(figs. 13 and 14). 

The division of the body cell just before fertilization results in 
unequal male cells (fig. 23), almost exactly resembling those of Taxus. 
It is not a case of the extrusion of one nucleus from the common 
cytoplasm, as observed by Coker (2) in Podocarpus, but the cyto- 
plasm is unequally divided, so that there are two distinct and naked 
male cells very different in size. The whole cavity of the pollen 
tube surrounding the cells and nuclei is rich in starch and other food 
materials (fig. 13). 

A consideration of the time involved in these various events shows 
that a period of about fifteen months elapses between the first appear- 
ance of the microsporangiate strobilus and fertilization, divided as 
follows: June, first appearance of strobilus; August, first appearance 
of stamineal primordia; September, distinct differentiation of spor- 
angia; April, shedding of spores; August, fertilization. The stro- 
bilus was first observed in July, but in such a condition that it must 
have been evident in June, if not earlier. In comparing this schedule 
with that given by Miss Robertson (8) for T. calijomica, it is evident 


that the sporangium in both species passes the winter in the mother- 
cell stage, but that the subsequent stages appear earlier in T. taxijolia 
growing in its natural habitat than in T. califomica growing in Eng- 
land. For example, in the latter case the reduction division takes 
place about the time that T. taxijolia sheds its pollen, and pollination 
in T. califomica does not seem to occur until late in May or early in 


Strobili. — The ovulate strobili are borne in the axils of the lower 
leaves of short young shoots (fig. 2). They usually occur in pairs 
upon a very short axillary branch, usually one pair, frequently two 
pairs, very rarely three pairs appearing upon a single shoot. A 
cluster of two to six strobili, therefore, appears near the base of the 
strobilus-bearing shoot, the upper pairs never maturing, usually 
one and sometimes both strobili of the lowest pair producing the 
large plum-like seeds (fig. j). The strobilus is a very simple one, 
consisting of four enveloping bracts and a single terminal ovule with 
two integuments (figs. 17 and 18), the outer one often called an arillus 
because it ripens fleshy. The whole structure resembles a simple 
ovulate flower with a perianth of four bracts, which perhaps deserve 
to be called a perianth as much as the so-called perianth of Gnetales; 
but it is none the less evident that they are the sterile bracts of a much 
reduced strobilus. 

The strobili were first seen July 26, at which time the growing 
point, enclosed by the bracts, was composed of entirely homogeneous 
tissue, showing no differentiation as an ovule (fig. 17). No subse- 
quent stages were found until April 7 (fig. 18), when the mother-cells 
were in synapsis (fig. 20) ; and it seems probable that the winter was 
passed in the mother-cell stage. At this time the integuments are 
entirely free from the nucellus, appearing to arise from the base of 
the ovule. Soon, however, extensive intercalary growth below the 
mother-cell becomes evident, the ovule being greatly elongated and 
broadened below, the original free nucellus with its integuments 
forming only the tip. This growth continues throughout the season 
of fertilization and the following one; and Oliver (6), who has 
described this intercalary growth in T. nucifera, estimates that in 
that species in the maturing seed the original nucellus with its free 


integuments represents only one- twentieth of the entire length of the 
seed. The characteristic structures of the integuments are con- 
tinued as two distinct peripheral layers of this large mass of additional 
tissue, and the tissue within these layers may be taken to represent 
in a similar way the downward extension of the nucellus. It is this 
additional nucellar tissue that the endosperm chiefly invades, giving 
rise to the phenomenon of " rumination," to be described later. 
Oliver (7) has also described fully the course of the vascular strands 
in the ovule of certain species of Torreya, a description which seems 
to serve as well for T. taxijolia. 

Female gametophyte. — The mother-cell is solitary and with 
no differentiation of a nutritive mechanism about it, such as appears 
in connection with the " spongy tissue" of Pinus, and in all the 
Pinaceae investigated. In Torreya it is directly in contact with the 
cells that are resorbed, without any intervening digestive layer (fig. 
20). Since this is true also of Podocarpus (2) and Taxus (10), it 
suggests the possibility that Taxaceae in general may be character- 
ized by the absence of a special digestive layer about the mother- cell. 

The reduction division was not seen, but a more or less complete 
tetrad is formed, as observed by Miss Robertson (8) also in T. 

The germination of the megaspore begins with the usual free 
nuclear division, the nuclei being in the parietal position when only 
sixteen to thirty- two in number (figs. 14 and 15). The interior of 
the sac contains cytoplasmic material, much less dense than the 
cytoplasm of the parietal layer, and also some reserve food. In this 
early few-nucleate stage of the endosperm there is always an appear- 
ance suggesting that the sac has sent a beak-like projection, contain- 
ing a nucleus, upwards into the nucellar tissue (fig. 15). After a 
careful comparison of the position of this apparent projection in 
reference to the surrounding parts with that of the megaspore, it 
seems that the " projection" is the original site of the megaspore, 
and that the appearance of a projection is due to the fact that the sac 
has encroached almost exclusively upon the chalazal tissue. This 
conspicuous beak, containing one of the parietal nuclei, often appears 
close to the tip of an advancing pollen tube, and suggests a possible 
explanation of the peculiar behavior ascribed to the archegonium 


initial of Tumboa. Inasmuch as the archegonium initial of Torreya 
often occupies this beak, the suggestion becomes still more pertinent 
(fig. 21). 

The formation of walls in the endosperm was not observed before 
July i, and in several instances they did not appear for a month later. 
When wall-formation began, repeated countings showed 256 free 
nuclei, which seems to be a very common limit of free nuclear division 
among gymnosperms. 

Since fertilization was observed August 12, it is evident that the 
archegonium is developed very early in the history of the gameto- 
phyte. In fact, as soon as the very small sac is filled with extremely 
delicate tissue the archegonium initial becomes evident. It does not 
seem possible for archegonia to appear any earlier, for the initials 
are organized as soon as the free nuclear stage has passed. In 
Torreva, therefore, nearly all of the endosperm, which becomes an 
extensive tissue, develops after fertilization. 

The single archegonium initial is always to one side of the central 
axis (fig. 16), often occupying the "beak" referred to above, and so 
projecting above the endosperm (fig. 21). A neck cell is cut off 
and divides anticlinally, forming a two-celled neck (fig. 22), the 
usual limit of neck-formation among gymnosperms. In fact, it is 
only among Podocarpeae and Abieteae that a more extensive neck 
is usually formed, consisting of more than one tier of cells, unless the 
somewhat anomalous neck of Ephedra be included. The fact that 
there is variation in the number of neck cells in the same form (two to 
twenty-five in Podocarpus), and that as a rule necks are destroyed 
as soon as formed by the growth of the central cell and pollen tube 
(fig. 23), suggests that their extent depends somewhat upon the 
approach of the pollen tube, which usually checks neck formation 
early, but sometimes permits it to become more extensive. In T. 
calijomica Miss Robertson (9) has found that the archegonia are 
usually three in number, ranging from two to five, and that the necks 
consist of four or six cells. 

The central cell enlarges rapidly, no jacket-layer being evident 
until after fertilization, and even then it is weakly organized (fig. 25). 
The nucleus is spherical and lies near but not against the neck cells, 
more nearly resembling an egg nucleus than in any gymnosperm 


we have observed (fig. 22). We could not detect the formation of a 
ventral canal cell or nucleus, or anything that stood for such a struct- 
ure at later stages. A ventral nucleus was expected, for a distinct 
ventral canal cell among Coniferales seems to be restricted to the 
Abieteae and does not always occur in them, and there seemed to be 
no excuse in our preparations for missing it. We are fully aware 
that all previous negative evidence as to the occurrence of at least a 
ventral nucleus in archegonium-forming gymnosperms has proved 
to be deceptive, but a study of the behavior of the central cell of 
Torreya, from the formation of the neck cell to fertilization, not only 
failed to show any indication of division but suggested that it may 
not occur. In T. californica a spindle seen twice in the central cell 
was interpreted by Miss Robertson (9) as representing the " cutting 
off" of a ventral nucleus, but no other traces of it could be found. 

In the single case in which two archegonia were observed, they 
were at opposite sides of the gametophyte, with the tip of a pollen 
tube between them. 

At the time of fertilization the gametophyte contains 400-800 
cells, with extremely thin walls and scanty cytoplasm (fig. 23). The 
only differentiation observable is the abundant accumulation of 
reserve food in the peripheral cells of the antipodal region. The 
whole mass of endosperm at this period usually measures 20 by 30/4 ; 
while in the mature seed the endosperm mass is ordinarily about 
20 mm i on g ky j^mm at j ts w id es t p ar t ? an d all of it surcharged with 
starch and other food materials. The food material is particularly 
conspicuous in a broad central band extending from the advancing 
tip of the embryo and widening to the antipodal end of the sac. Fig. 
5 shows the longitudinal extent of the band, and fig. 4 its cross-section. 
The peculiar behavior of the endosperm after fertilization will be 
considered under the discussion of the maturing of the seed. 


The forcible discharge of the contents of the tube may be inferred 
from the vacuole-like appearance in the center of the egg, produced 
by the inrush (figs. 24 and 25). The male nucleus in its cytoplasmic 
sheath passes through the cytoplasm of the egg and comes in contact 
with the egg nucleus. The male cytoplasm becomes closely appressed 


to the surface of the female nucleus, slips from its own nucleus, and 
was observed extending over fully two-thirds of the female nucleus 
(fig. 24). This behavior of the male cytoplasm has been observed 
by Coker (3) in Taxodium and by Miss Robertson (9) in Torreya 
calijomica. The male cytoplasm of Torreya taxifolia is sharply 
differentiated by staining from the cytoplasm of the egg, and undoubt- 
edly completely invests the fusion nucleus. The appearance of a 
similar mass of cytoplasm investing the free nuclei of the first division 
(fig. 25), and continued in the second division in connection with 
wall-formation (fig. 26) suggests the possibility that the male cyto- 
plasm may remain differentiated through more than one cell genera- 
tion. Near the neck end of the archegonium nuclei are evident, 
which seem to be the stalk and tube nuclei and the other male nucleus 
with its investing cytoplasm (fig. 24). 


Soon after fertilization the first division of the egg nucleus was 
observed (fig. 25), and almost immediately the second division fol- 
lows, giving rise to four large free nuclei almost filling the egg (fig. 
26), one nucleus in the base of the egg, the other three in a plane 
above. At this time wall-formation occurs, the cytoplasmic radiations 
which precede it being very evident (fig. 26). Two weeks later the 
egg is completely filled with a proembryo consisting of twelve to 
eighteen cells (fig. 27). This complete filling of the egg by the pro- 
embryo is remarkable among Coniferales, having recently been 
observed also by Lawson (4) in Sequoia, but as yet not recorded 
in other genera. In Torreya, at least, this fact seems to be related 
to the relatively small size of the egg, the very large nuclei, and the 
early appearance of walls. 

The cells of the proembryo at this early stage are distinctly in three 
tiers; that nearest the neck of the archegonium comprising five or six 
cells and forming the primary suspensor tier; the middle tier, com- 
prising five or six cells and forming the secondary suspensor tier; and 
the lowest consisting of a single cell which ultimately contributes to 
suspensor-formation and forms the embryo. The inequality in the 
number of cells entering into the tiers seems to be characteristic of 
Taxaceae. In Podocarpus Coker (2) found the three tiers made up 


of eleven cells in each of the upper two and one in the lowest; and a 
similar but less striking inequality is to be observed in Taxus. Occa- 
sionally in Torreya other divisions may occur, giving rise to approxi- 
mately four tiers and a proembryo of about eighteen cells; but no 
other division occurs until the following spring, the winter condition 
being a proembryo of three or four tiers of cells as described above. 

In the following season the suspensor develops and the embryo 
is formed, along with the characteristic " rumination" of the endo- 
sperm and the development of the testa. The first indication of 
change from the winter proembryonic condition is the elongation of 
the primary suspensor cells (fig. 29); a little later this is shared by 
the secondary suspensor cells; and this is followed by the elongation 
of the third tier, if four tiers are formed. In the meantime the termi- 
nal cell has begun a series of rapid divisions, resulting in a cylindrical 
mass of meristematic tissue, much as Lyon (5) has described in the 
case of the embryo of Ginkgo. This meristematic cylinder advances 
gradually into the endosperm, its basal tiers of cells successively 
contributing to the suspensor elongation. Thus in the formation 
of the suspensor there seems to be developed what may be called a 
wave of elongation, beginning with the uppermost tier of the pro- 
embryo and extending gradually downward, tier after tier, until it 
includes the upper region of the meristematic cylinder formed by 
the terminal cell. This same phenomenon is very evident also in 

After the meristematic cylinder has advanced into the endosperm 
and has become prominent, the growing points are organized; the 
two cotyledons presently becoming beautifully crescentic in outline 
and completely surrounding the stem- tip; and the root-tip being 
organized deep within the meristematic cylinder. 

In several instances a number of small embryos were observed 
imbedded in the endosperm about the suspensor region of the normal 
embryo. Our material did not permit any determination of their 
origin, but they resemble the proembryo of the normal embryo, and 
are developed while the latter is in its second season's growth. After 
the pollen tube has reached the archegonium, the endosperm grows 
up around it (figs. 23 and 24), so that the tube lies in a cup-shaped 
depression. After fertilization, the rim of this endosperm cup con- 


tinues growth and gradually incloses trie fertilized egg, in most cases 
forming quite an elongated beak above the embryo. Later many 
of these cells round off, forming a loose tissue, and it is among these 
rounded-off cells that the feeble accessory embryos are produced. 
Whether these have been developed apogamously from the endosperm 
cells, or have budded from the suspensor cells can only be conjectured. 
In any event, they might develop further if there was any failure in 
the development of the normal embryo. 

The time involved in the series described above, that is from the 
first appearance of the megasporangiate strobilus to the maturity of 
the seed, is about thirty months, distributed as follows: June (?), 
first appearance of the strobilus; April, mother- cells in synapsis; 
August, fertilization; October, proembryo of 12 to 18 cells (winter 
condition); following season, development of embryo, " rumination" 
of endosperm, and development of testa; October, fall of seed. 


The outer integument and its histological continuation about the 
ovule develops a thick fleshy coat containing very numerous large 
resin cavities, and completely inclosing the structures within (figs. 
4 and 5). This fleshy coat gives to the mature seed the appearance 
of a plum (fig. 3), as in the seeds of Cycads and Ginkgo. Within 
the broad band of resin cavities, near the inner limit of the integu- 
ment, two conspicuous vascular strands occur, directly opposite one 
another (fig. 4). These are the main strands of the very character- 
istic vascular system of the ovule described by Oliver (7). 

The inner integument early differentiates into two distinct layers, 
a differentiation just as evident in Cycads and Ginkgo. The outer 
layer forms the stony coat, and the transformation from soft to very 
hard tissue begins after the embryo and endosperm have completed 
their development. The hardening begins at the apex of the ovule, 
and on account of resistance to stains appears under low power as a 
clear band (black in fig. 5). The hardening band gradually extends 
downwards through the relatively very short integument, and differ- 
entiating as a distinct layer in the much larger mass of tissue below 
completely invests the ovule within the fleshy coat. Protoplasmic 
connections between the cells of the stony coat and striations in the 
cell walls are unusually clear. 


The inner layer of the inner integument comprises several layers 
of thin- walled cells, but beyond the integuments it is not histologically 
differentiated as a layer distinct from the tissue within. Accordingly, 
through the great bulk of the seed this layer may be neglected, and 
the whole mass of tissue within the stony layer and outside of the 
embryo sac will be spoken of as nuclear tissue or perisperm. 

The behavior of the endosperm is peculiar, resulting in what is 
called " ruminated endosperm," a phenomenon peculiar to Torreya 
among gymnosperms, and commonly illustrated by the nutmeg 
" Rumination" of endosperm proves to be a misnomer, for the endo- 
sperm is always the successfully aggressive tissue in developing this 
condition. The perisperm continues to grow throughout the matur- 
ing of the seed, and the final condition results from what might be 
called the struggle of two growing tissues that have been abutting upon 
one another through their whole period of growth. 

In ordinary seeds the endosperm invades the surrounding tissue 
more or less uniformly; in the case of Cycads and Ginkgo, for exam- 
ple, obliterating most of the perisperm. In Torreya, on the other 
hand, the invasion by the endosperm is irregular in the extreme. 
It is in the season after the proembryo has been formed that the active 
invasion of the perisperm begins. The extension of the endosperm 
into the tip of the nucellus above the sac proceeds in the usual way, 
obliterating all of it except a few peripheral layers of cells. This uni- 
form invasion seems to be due to the fact that in this apical region 
(the original nucellus) the perisperm is not growing actively if at all. 
Below this small region at the tip, however, the perisperm is very active 
and evidently resists disintegration much more at certain points than 
at others. As a consequence, the perisperm becomes eroded by the 
irregularly advancing endosperm, and is left in the condition of a much 
dissected coast-line (figs. 4 and 5). To the casual observer this results 
in an appearance suggesting that the endosperm is being invaded by 
plates of perisperm, but this is no more true than that the promontories 
of a dissected coast-line are advancing into the sea. The suggestion of 
an invading perisperm is further strengthened by the fact that within 
the perisperm bordering the endosperm a dark brown and finally 
black band of cells is developed, due to abundant food storage (figs. 
4, 30, 31), but this really recedes as the endosperm advances. 


A cross-section of the mature seed always shows a definite and deep 
constriction of the endosperm in the center, exactly opposite the two 
opposed vascular strands that run up on each side of the seed through 
the inner part of the outer integument (fig. 4). This constriction is 
the cross-section of two opposite and deep longitudinal furrows in the 
endosperm, and it means that in this longitudinal plane the endosperm 
encounters the greatest resistance in invading the perisperm. This 
most resistant perisperm certainly seems to hold a very definite topo- 
graphical relation to the principal vascular strands, and this relation 
may explain the resistance. 

That endosperm is the aggressive tissue at every point, even at 
the region of most resistant perisperm, is evident for several reasons. 
In no case were the peripheral cells of the endosperm broken down; 
and in no case did there fail to appear one or two layers of disorgan- 
ized cells of the perisperm in contact with the endosperm (figs. 30 
and 31). In every case, also, the peripheral cells of the endosperm 
appeared active and very vigorous, and their different appearance in 
regions of more and less active encroachment is striking. In regions 
of active invasion the endosperm cells are radially elongated, and 
many of them are binucleate (fig. 30) ; while in regions of less active 
invasion the cells are more nearly isodiametric and rarely binucleate 

(fig- 3*)- 

Another proof that endosperm is the encroaching tissue may be 
obtained from comparative measurements. At the times of fertili- 
zation the gametophyte usually measures 20 by 30/x. In the mature 
seed the ordinary length of the gametophyte is 20 mm , the greatest 
width being i4 mm , and the least (at the deep constriction opposite 
the vascular strands)) i.5 mm . At this most resistant region of the 
perisperm, therefore, where it is hard to escape the conviction that 
the perisperm plate has cut the endosperm nearly in two, the endo- 
sperm has increased its diameter against the perisperm seventy-five 

The best reason, however, for concluding that the endosperm is 
the invading tissue in this case is that this is always the behavior of 
endosperm; and it is singular that the old explanation of " rumination' ' 
was ever suggested. An examination of the nutmeg, the classic 
illustration of " ruminated endosperm/ ' and of Asimina triloba, 


showed that precisely the same explanation applies to them that we 
have given in the case of Torreya. 


The staminate strobilus consists of a series of closely overlapping 
sterile bracts, in four vertical rows, completely enveloping the tip of 
the axis bearing numerous stamens. The large adaxial resin cavity 
that occurs in the stamen occupies the site of three abortive sporangia. 

The male gametophyte has no prothallial cell, and the male cells 
are very unequal, resembling those of Taxus. The pollen tube is 
exceedingly variable in the rate and direction of its advance through 
the nucellar cap, sometimes pushing in the embryo sac while it is 
in an early free-nucleate stage. 

The ovulate strobilus consists of four enveloping bracts and a 
single terminal ovule with two integuments. Extensive intercalary 
growth below the mother-cell forms the bulk of the mature ovule 
and seed. There is no organization of a special digestive layer 
around the mother-cell. 

The solitary archegonium initial appears as soon as walls are 
formed, is always at one side of the central axis of the gametophyte, 
and forms a two-celled neck. The nucleus of the central cell was 
not observed to divide, nor could any trace of a ventral nucleus be 

In fertilization the male cytoplasm invests the fusion nucleus, 
and seems to remain distinct until wall-formation at the four- 
nucleate stage of the proembryo. 

In the development of the proembryo, four free nuclei appear 
before wall-formation, and the proembryo completely fills the egg, 
having no "open cells.' ' A proembryo of twelve to eighteen cells 
is the winter stage. In the spring the suspensor is formed by what 
may be called a wave of elongation, beginning with the uppermost 
tier of the proembryo, and extending gradually downward, tier after 
tier, until it includes the upper region of the meristematic cylinder 
formed by the terminal cell. 

Small embryos are formed during the second season in the sus- 
pensor region of the normal embryo; but whether they arise from 
prothallial or suspensor cells was not determined. 


The "rumination" of endosperm, peculiar to Torreya among 
gymnosperms, arises from the extremely irregular encroachment of 
the endosperm upon the perisperm, the endosperm being resisted 
much more at certain points than at others. The same was found 
to be true of other "ruminated" seeds. 

The University of Chicago. 


1. Chapman, A. W., Torreya taxijolia Arnott; a reminiscence. Bot. Gazette 

10:251-254. map. 1885. 

2. Coker, W. C, Notes on the gametophytes and embryo of Podocarpus. 

Bot. Gazette 33:89-107. pis. 5-7. 1902. 

3. , On the gametophytes and embryo of Taxodium. Bot. Gazette 

36:1-27, 114-140. pis. i-ii. 1903. 

4. Lawson, A. A., The gametophytes, archegonia, fertilization, and embryo of 

Sequoia sempervirens. Ann. Botany 18:1-28. pis. 1-4. 1904. 

5. Lyon, H. L., The embryogeny of Ginkgo. Minn. Bot. Studies 3:275-290. 

pis. 29-43. 1904. 

6. Oliver, F. W., On some points of apparent resemblance in certain fossil 

and recent gymnospermous seeds. New Phytologist 1 : 145-154. 1902. 

7. , The ovules of the older gymnosperms. Ann. Botany 17:451-476. 

pi. 24. 1903. 

8. Robertson, Agnes, Spore formation in Torreya calif ornica. New Phytolo- 

gist 3:133-148. pis. 3~4- 1904. 

9. — , Studies in the morphology of Torreya calif ornica Torr. II. The 

sexual organs and fertilization. New Phytologist 3:205-216. pis. y-g. 
10. Strasburger, E., Anlage des Embryosackes und Prothalliumbildung bei 
der Eibe nebst anschliessenden Erorterungen. Festschrift Ernst Haeckel. 
Jena. 1904. 


With the exception of Plate 7, all figures were drawn with the aid of an Abbe 
camera lucida and reduced one-half in reproduction. Abbreviations are as 
follows: br, bract; a, primordium of stamen; oi, outer integument; ii, inner 
integument; r, resin cavity; n, nucellus; pc, pollen chamber; s, sporogenous 
cells; tn, tube nucleus; g, primary spermatogenous cell; stn, nucleus of stalk 
cell; b, body cell; Wi functional male cell; m 2 functionless male cell; hr, haus- 
torial region of female gametophyte; nc, nucleus of neck cell; 0, egg; end, endo- 
sperm; per, perisperm. 



Fig. 1. Staminate branch; October 21, 1904. Xi. 

Fig. 2. Ovulate branch; April 7, 1904. Xi. 

Fig. 3. Mature seed; October 21, 1904. Xi. 

Fig. 4. Cross-section of seed showing ruminated endosperm with storage 
region in center, seed coats, stored food in the perisperm, and resin ducts. X3. 

Fig. 5. Longitudinal median section through seed showing embryo, rumi- 
nated endosperm, and seed coats. X3. 


Fig. 6. Transverse section through a staminate strobilus showing bracts and 
primordia of stamens; August 12, 1904. X24. 

Fig. 7. Longitudinal section through a staminate strobilus showing envelop- 
ing bracts, primordia of stamens, and storage region; the position of the bundles 
is shown by the dotted lines; August 12, 1904. X24. 

Fig. 8. Primordium of a stamen; August 12, 1904. X460. 

Fig. 9. Median section through an adaxial sporangium showing the disor- 
ganizing primary sporogenous cell and two wall cells; October 21, 1904. X460. 

Fig. 10. Cross-section of a sporophyll showing the three abortive adaxial 
and the three functional abaxial sporangia; October 21, 1904. X47. 

Fig. 11. Longitudinal section through a stamen showing early stage of the 
resin cavity above and a functioning sporangium below; October 21, 1904. X 255. 

Fig. 12. Pollen grain showing primary spermatogenous cell and tube cell; 
April 7, 1904. X1250. 

Fig. 13. Tip of a pollen tube which has penetrated about half way through 
the nucellus; June 10, 1904. X460. 

Fig. 14. Tip of pollen tube in contact with the embryo sac; the female garnet - 
ophyte is in the free nuclear stage and the nuclei are placed parietally; June 21, 
1904. X460. 

Fig. 15. Embryo sac in free nuclear stage showing one of the nuclei occupy- 
ing the place of the megaspore mother-cell; June 10, 1904. X185. 

Fig. 16. Pollen tube in contact with the endosperm; an archegonium is at 
the left of the tube; July 26, 1904. X255. 

PLATE 11. 

Fig. 17. Young ovulate strobilus showing enveloping bracts, inner integu- 
ment, and nucellus; July 26, 1904. X47. 

Fig. 18. Ovulate strobilus showing bracts, integuments, nucellus with rudi- 
mentary pollen chamber, and megaspore mother-cell; April 7, 1904. X24. 

Fig. 19. Tip of nucellus with pollen chamber containing a microspore; 
April 8, 1904. X220. 

Fig. 20. Megaspore mother-cell in synapsis; April 7, 1904. X460. 

Fig. 21. Micropylar end of female gametophyte with archegonium initial 
projecting into the space formerly occupied by the megaspore mother-cell; July 
26, 1904. X460. 


Fig. 22. Archegonium consisting of two neck cells and central cell; the 
nucleus of the central cell has rounded out, and is passing downward to the center 
of the cell and taking on the appearance of an egg nucleus; July 26, 1904. X460. 

Fig. 23. Median longitudinal section of female gametophyte showing egg, 
remains of a neck cell, and antipodal haustorial cells; the tip of the pollen tube 
in contact with the egg contains stalk and tube nuclei, functional and functionless 
male cells; the upward growth of the endosperm cells forms a sheath around 
the pollen tube; August 12, 1904. X460. 

PLATE ill. 

Fig. 24. Fertilization; the male nucleus is in contact with the egg nucleus; 
the cytoplasm of the male cell is closely applied to the egg nucleus; the function- 
less male cell, and tube and stalk nuclei are in the upper part of the egg cytoplasm; 
the cavity in the egg cytoplasm is caused by the inrush of the contents of the 
pollen tube; August 12, 1904. X460. 

Fig. 25. Two-celled proembryo; the dense cytoplasmic mass surrounding 
the nuclei is probably in greater part derived from the male cytoplasm; August 
27, 1904. X460. 

Fig. 26. Four-celled proembryo; walls coming in; August 12, 1904. X460. 

Fig. 27. Proembryo shortly after walls are laid down; the proembryo passes 
the winter in this stage; August 27, 1904. X460. 

Fig. 28. Cross-section through suspensor cells; September 12, 1904. X460. 

Fig. 29. Proembryo showing elongation of suspensor cells; April 7, 1904. 

Fig. 30. Endosperm cells encroaching on perisperm; August 12, 1904. X640. 

Fig. 31. Endosperm cells which have ceased to encroach on perisperm; 
August 12, 1904. X460. 



UU(*V. pY*l 




VJ0.3.C*.Wd.. A*L. 



\&3> £.Sui»mL, ^UL.